simplecart.java

代码是一个分类器的实现,其中使用了部分weka的源代码。可以将项目导入eclipse运行

    return "CART Decision Tree\n" + toString(0)+"\n\n"
    +"Number of Leaf Nodes: "+numLeaves()+"\n\n" +
    "Size of the Tree: "+numNodes();
  }

  /**
   * Outputs a tree at a certain level.
   * 
   * @param level 	the level at which the tree is to be printed
   * @return 		a tree at a certain level
   */
  protected String toString(int level) {

    StringBuffer text = new StringBuffer();
    // if leaf nodes
    if (m_Attribute == null) {
      if (Instance.isMissingValue(m_ClassValue)) {
	text.append(": null");
      } else {
	double correctNum = (int)(m_Distribution[Utils.maxIndex(m_Distribution)]*100)/
	100.0;
	double wrongNum = (int)((Utils.sum(m_Distribution) -
	    m_Distribution[Utils.maxIndex(m_Distribution)])*100)/100.0;
	String str = "("  + correctNum + "/" + wrongNum + ")";
	text.append(": " + m_ClassAttribute.value((int) m_ClassValue)+ str);
      }
    } else {
      for (int j = 0; j < 2; j++) {
	text.append("\n");
	for (int i = 0; i < level; i++) {
	  text.append("|  ");
	}
	if (j==0) {
	  if (m_Attribute.isNumeric())
	    text.append(m_Attribute.name() + " < " + m_SplitValue);
	  else
	    text.append(m_Attribute.name() + "=" + m_SplitString);
	} else {
	  if (m_Attribute.isNumeric())
	    text.append(m_Attribute.name() + " >= " + m_SplitValue);
	  else
	    text.append(m_Attribute.name() + "!=" + m_SplitString);
	}
	text.append(m_Successors[j].toString(level + 1));
      }
    }
    return text.toString();
  }

  /**
   * Compute size of the tree.
   * 
   * @return 		size of the tree
   */
  public int numNodes() {
    if (m_isLeaf) {
      return 1;
    } else {
      int size =1;
      for (int i=0;i<m_Successors.length;i++) {
	size+=m_Successors[i].numNodes();
      }
      return size;
    }
  }

  /**
   * Method to count the number of inner nodes in the tree.
   * 
   * @return 		the number of inner nodes
   */
  public int numInnerNodes(){
    if (m_Attribute==null) return 0;
    int numNodes = 1;
    for (int i = 0; i < m_Successors.length; i++)
      numNodes += m_Successors[i].numInnerNodes();
    return numNodes;
  }

  /**
   * Return a list of all inner nodes in the tree.
   * 
   * @return 		the list of all inner nodes
   */
  protected Vector getInnerNodes(){
    Vector nodeList = new Vector();
    fillInnerNodes(nodeList);
    return nodeList;
  }

  /**
   * Fills a list with all inner nodes in the tree.
   * 
   * @param nodeList 	the list to be filled
   */
  protected void fillInnerNodes(Vector nodeList) {
    if (!m_isLeaf) {
      nodeList.add(this);
      for (int i = 0; i < m_Successors.length; i++)
	m_Successors[i].fillInnerNodes(nodeList);
    }
  }

  /**
   * Compute number of leaf nodes.
   * 
   * @return 		number of leaf nodes
   */
  public int numLeaves() {
    if (m_isLeaf) return 1;
    else {
      int size=0;
      for (int i=0;i<m_Successors.length;i++) {
	size+=m_Successors[i].numLeaves();
      }
      return size;
    }
  }

  /**
   * Returns an enumeration describing the available options.
   *
   * @return 		an enumeration of all the available options.
   */
  public Enumeration listOptions() {
    Vector 	result;
    Enumeration	en;
    
    result = new Vector();
    
    en = super.listOptions();
    while (en.hasMoreElements())
      result.addElement(en.nextElement());

    result.addElement(new Option(
	"\tThe minimal number of instances at the terminal nodes.\n" 
	+ "\t(default 2)",
	"M", 1, "-M <min no>"));
    
    result.addElement(new Option(
	"\tThe number of folds used in the minimal cost-complexity pruning.\n"
	+ "\t(default 5)",
	"N", 1, "-N <num folds>"));
    
    result.addElement(new Option(
	"\tDon't use the minimal cost-complexity pruning.\n"
	+ "\t(default yes).",
	"U", 0, "-U"));
    
    result.addElement(new Option(
	"\tDon't use the heuristic method for binary split.\n"
	+ "\t(default true).",
	"H", 0, "-H"));
    
    result.addElement(new Option(
	"\tUse 1 SE rule to make pruning decision.\n"
	+ "\t(default no).",
	"A", 0, "-A"));
    
    result.addElement(new Option(
	"\tPercentage of training data size (0-1].\n" 
	+ "\t(default 1).",
	"C", 1, "-C"));

    return result.elements();
  }

  /**
   * Parses a given list of options. <p/>
   * 
   <!-- options-start -->
   * Valid options are: <p/>
   * 
   * <pre> -S &lt;num&gt;
   *  Random number seed.
   *  (default 1)</pre>
   * 
   * <pre> -D
   *  If set, classifier is run in debug mode and
   *  may output additional info to the console</pre>
   * 
   * <pre> -M &lt;min no&gt;
   *  The minimal number of instances at the terminal nodes.
   *  (default 2)</pre>
   * 
   * <pre> -N &lt;num folds&gt;
   *  The number of folds used in the minimal cost-complexity pruning.
   *  (default 5)</pre>
   * 
   * <pre> -U
   *  Don't use the minimal cost-complexity pruning.
   *  (default yes).</pre>
   * 
   * <pre> -H
   *  Don't use the heuristic method for binary split.
   *  (default true).</pre>
   * 
   * <pre> -A
   *  Use 1 SE rule to make pruning decision.
   *  (default no).</pre>
   * 
   * <pre> -C
   *  Percentage of training data size (0-1].
   *  (default 1).</pre>
   * 
   <!-- options-end -->
   * 
   * @param options the list of options as an array of strings
   * @throws Exception if an options is not supported
   */
  public void setOptions(String[] options) throws Exception {
    String	tmpStr;
    
    super.setOptions(options);
    
    tmpStr = Utils.getOption('M', options);
    if (tmpStr.length() != 0)
      setMinNumObj(Double.parseDouble(tmpStr));
    else
      setMinNumObj(2);

    tmpStr = Utils.getOption('N', options);
    if (tmpStr.length()!=0)
      setNumFoldsPruning(Integer.parseInt(tmpStr));
    else
      setNumFoldsPruning(5);

    setUsePrune(!Utils.getFlag('U',options));
    setHeuristic(!Utils.getFlag('H',options));
    setUseOneSE(Utils.getFlag('A',options));

    tmpStr = Utils.getOption('C', options);
    if (tmpStr.length()!=0)
      setSizePer(Double.parseDouble(tmpStr));
    else
      setSizePer(1);

    Utils.checkForRemainingOptions(options);
  }

  /**
   * Gets the current settings of the classifier.
   * 
   * @return 		the current setting of the classifier
   */
  public String[] getOptions() {
    int       	i;
    Vector    	result;
    String[]  	options;

    result = new Vector();

    options = super.getOptions();
    for (i = 0; i < options.length; i++)
      result.add(options[i]);

    result.add("-M");
    result.add("" + getMinNumObj());
    
    result.add("-N");
    result.add("" + getNumFoldsPruning());
    
    if (!getUsePrune())
      result.add("-U");
    
    if (!getHeuristic())
      result.add("-H");
    
    if (getUseOneSE())
      result.add("-A");
    
    result.add("-C");
    result.add("" + getSizePer());

    return (String[]) result.toArray(new String[result.size()]);	  
  }

  /**
   * Return an enumeration of the measure names.
   * 
   * @return 		an enumeration of the measure names
   */
  public Enumeration enumerateMeasures() {
    Vector result = new Vector();
    
    result.addElement("measureTreeSize");
    
    return result.elements();
  }

  /**
   * Return number of tree size.
   * 
   * @return 		number of tree size
   */
  public double measureTreeSize() {
    return numNodes();
  }

  /**
   * Returns the value of the named measure.
   * 
   * @param additionalMeasureName 	the name of the measure to query for its value
   * @return 				the value of the named measure
   * @throws IllegalArgumentException 	if the named measure is not supported
   */
  public double getMeasure(String additionalMeasureName) {
    if (additionalMeasureName.compareToIgnoreCase("measureTreeSize") == 0) {
      return measureTreeSize();
    } else {
      throw new IllegalArgumentException(additionalMeasureName
	  + " not supported (Cart pruning)");
    }
  }

  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui
   */
  public String minNumObjTipText() {
    return "The minimal number of observations at the terminal nodes (default 2).";
  }

  /**
   * Set minimal number of instances at the terminal nodes.
   * 
   * @param value 	minimal number of instances at the terminal nodes
   */
  public void setMinNumObj(double value) {
    m_minNumObj = value;
  }

  /**
   * Get minimal number of instances at the terminal nodes.
   * 
   * @return 		minimal number of instances at the terminal nodes
   */
  public double getMinNumObj() {
    return m_minNumObj;
  }

  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui
   */
  public String numFoldsPruningTipText() {
    return "The number of folds in the internal cross-validation (default 5).";
  }

  /** 
   * Set number of folds in internal cross-validation.
   * 
   * @param value 	number of folds in internal cross-validation.
   */
  public void setNumFoldsPruning(int value) {
    m_numFoldsPruning = value;
  }

  /**
   * Set number of folds in internal cross-validation.
   * 
   * @return 		number of folds in internal cross-validation.
   */
  public int getNumFoldsPruning() {
    return m_numFoldsPruning;
  }

  /**
   * Return the tip text for this property
   * 
   * @return 		tip text for this property suitable for displaying in 
   * 			the explorer/experimenter gui.
   */
  public String usePruneTipText() {
    return "Use minimal cost-complexity pruning (default yes).";
  }

  /** 
   * Set if use minimal cost-complexity pruning.
   * 
   * @param value 	if use minimal cost-complexity pruning
   */
  public void setUsePrune(boolean value) {
    m_Prune = value;
  }

  /** 
   * Get if use minimal cost-complexity pruning.
   * 
   * @return 		if use minimal cost-complexity pruning
   */
  public boolean getUsePrune() {
    return m_Prune;
  }

  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui.
   */
  public String heuristicTipText() {
    return 
        "If heuristic search is used for binary split for nominal attributes "
      + "in multi-class problems (default yes).";
  }

  /**
   * Set if use heuristic search for nominal attributes in multi-class problems.
   * 
   * @param value 	if use heuristic search for nominal attributes in 
   * 			multi-class problems
   */
  public void setHeuristic(boolean value) {
    m_Heuristic = value;
  }

  /** 
   * Get if use heuristic search for nominal attributes in multi-class problems.
   * 
   * @return 		if use heuristic search for nominal attributes in 
   * 			multi-class problems
   */
  public boolean getHeuristic() {return m_Heuristic;}

  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui.
   */
  public String useOneSETipText() {
    return "Use the 1SE rule to make pruning decisoin.";
  }

  /** 
   * Set if use the 1SE rule to choose final model.
   * 
   * @param value 	if use the 1SE rule to choose final model
   */
  public void setUseOneSE(boolean value) {
    m_UseOneSE = value;
  }

  /**
   * Get if use the 1SE rule to choose final model.
   * 
   * @return 		if use the 1SE rule to choose final model
   */
  public boolean getUseOneSE() {
    return m_UseOneSE;
  }

  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui.
   */
  public String sizePerTipText() {
    return "The percentage of the training set size (0-1, 0 not included).";
  }

  /** 
   * Set training set size.
   * 
   * @param value 	training set size
   */  
  public void setSizePer(double value) {
    if ((value <= 0) || (value > 1))
      System.err.println(
	  "The percentage of the training set size must be in range 0 to 1 "
	  + "(0 not included) - ignored!");
    else
      m_SizePer = value;
  }

  /**
   * Get training set size.
   * 
   * @return 		training set size
   */
  public double getSizePer() {
    return m_SizePer;
  }

  /**
   * Main method.
   * @param args the options for the classifier
   */
  public static void main(String[] args) {
    runClassifier(new SimpleCart(), args);
  }
}